IDENTIFICATION OF T-CELL AUTOEPITOPES THAT CROSS-REACT WITH THE C-TERMINAL SEGMENT OF THE M-PROTEIN OF GROUP-A STREPTOCOCCI

Rheumatic fever (RF) follows a throat infection with different M-serot ypes of beta-hemolytic group A streptococci (GAS) and can affect diffe rent tissues, predominantly the heart. It is thought to be an autoimmu ne illness. Although histological examination of affected heart shows an infiltrate consisting mainly of T cells, antigens or epitopes that could be putative targets of autoimmune T cells have not been identifi ed. We have examined the T cell response to the conserved C-terminal r egion of the M protein-a streptococcal surface coiled-coil protein whi ch is the target of opsonic antibodies and antibodies which cross-reac t with human heart tissue. Australian Aborigine, Caucasian and Thai pa tients, controls and mice were studied to define regions of the protei n immunogenic for T cells, and T cell lines and clones were tested for cross-reactivity to myosin as well as an extract of Rf-diseased mitra l heart valve. Murine (B10, B10.D2, B10.BR) M peptide-specific T cells were often cross-reactive for other M peptides but did not cross-reac t with human heart antigens. Patients with RF or other heart diseases, or control subjects exposed more commonly to GAS were more likely to have T cell responses to the M protein, with many regions of the C-ter minus being recognized. T cell lines and a clone specific for differen t M peptides were generated from five donors. Cross-reactivity could b e shown between different M peptides, but unlike murine M peptide-spec ific T cells three of the human T cell lines reacted strongly to pepti des representing homologous regions of cardiac and skeletal muscle myo sins, and two of these lines also responded to porcine myosin and an e xtract of human rheumatic mitral valve. However, these last two lines were derived from a normal donor without history of RF or other heart disease. Our data demonstrate that regions of the M protein, including regions that are being considered as subunit vaccines, have the poten tial to stimulate pre-existing heart cross-reactive T cells, but that the ability of such T cells to cross-react (as measured in vitro) is n ot in itself sufficient to lead to disease.